© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-4B
    5. 20-TRANSCODER-HP
    7. 107419
    Prev
    Next
    INDEX
    Explanations

    snippets of text

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_20.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.20.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    adians
    -0.29
     Peters
    -0.27
    æĬ±
    -0.26
     discrepan
    -0.25
     Trib
    -0.24
    sei
    -0.24
    澤
    -0.24
     ##↵
    -0.24
    adian
    -0.23
     discrepancy
    -0.23
    POSITIVE LOGITS
    å¼Ī
    0.29
    agina
    0.25
    iti
    0.25
     конкÑĢеÑĤ
    0.24
    åłĥ
    0.23
    ãģĦãĤĦ
    0.23
    ÙĦاÙħ
    0.23
    еÑĪÑĮ
    0.23
     عÙĨÙĩ
    0.23
    itous
    0.23
    Activations Density 7.817%

    No Known Activations